This project began with the discovery in our laboratory of a new possible "modulatory" combination therapy for AIDS (AZT or ddC plus an inhibitor of the transport of physiological nucleosides into cells). One such inhibitor is dipyridamole, a commonly used cardiovascular agent. After studies of mechanism, molecular biology, and preclinical pharmacology, the combination of AZT and dipyridamole was carried through Phase I clinical trials with collaborators at the Henry Jackson Foundation and Johns Hopkins University. Phase I goals were met with respect to dose-ranging and clinical pharmacology. A phase II clinical protocol received all necessary approvals but was not carried out because of funding issues and job changes by key personnel. Our current work on combination therapy was an outgrowth of those studies. We could not find any published method for designing and analyzing experiments on combination therapy that appeared adequate to the task. Hence, we have developed our own. The result is a growing set of new concepts, algorithms, and computer programs, embedded in a package called COMBO. That work, which has general application to experiments on therapy of cancer and AIDS, now forms the central portion of this project, albeit a small fraction of overall effort in our research group. We have used the new algorithms and programs to design and analyze experiments on such agents as suramin, tumor necrosis factor, taxol, dipyridamole, AZT, ddI, ddC, CD4-pseudomonas exotoxin, HIV protease inhibitors, and hydroxyurea, among others. Analyses in the last year have been done on combinations of AZT or ddI with protease inhibitor KNI-272 with H. Mitsuya and colleagues (AIDS Research and Human Retroviruses 1994;38:1036) and with Gallo and colleagues (Science 1994;266:8Ol).